Explarartory Questions:¶

Properties and Pricing¶

How do neighborhoods differ in price?
Does proximity to attractions play a factor in price?
What kind of properties are these listings (property type, bedroom count, bathroom count, etc.)

Host "business" analysis: Are there hosts that have multiple listings/have businesses using Aribnb? Do these hosts have good scores and reviews? Do they tend to be in certain neighbourhoods and have listings with certain prices?

Where and maybe which host should I prefer to stay at/with to increase my changes of having a good experience? In both cheap and expensive neighbourhoods?

</ol>

# imports 
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import plotly.express as px
import seaborn as sns
import json
import ipywidgets as widgets
import plotly
%matplotlib inline

# my favorite
plt.style.use("fivethirtyeight")

# show full columns
pd.set_option('display.max_columns', None)

# cell width 
from IPython.core.display import display, HTML
display(HTML("<style>.container { width:80% !important; }</style>"))

Datasets available to us: listings, reviews, geographical information¶

# listings data
ls = pd.read_csv("../data/listings.csv")
ls_d = pd.read_csv("../data/listings 2.csv")

# reviews data
rs = pd.read_csv("../data/reviews.csv")
rs_d = pd.read_csv("../data/reviews 2.csv")


# geography data
geo = pd.read_csv("../data/neighbourhoods.csv")

with open("../data/neighbourhoods.geojson") as jsonfile:
    geojson = json.load(jsonfile)

/Users/ldugom/.local/lib/python3.6/site-packages/IPython/core/interactiveshell.py:3063: DtypeWarning:

Columns (61,62,94) have mixed types.Specify dtype option on import or set low_memory=False.

Preliminary: What/Where are the neighbourhoods in DC?¶

import plotly.express as px

# set token
px.set_mapbox_access_token("pk.eyJ1IjoibGF3cmVuY2VkIiwiYSI6ImNrODFzZnFnNzA0YmczZW9nNWN4aTFvdngifQ.VlB5-L7owXKEXo8JEePk7w")
fig = px.choropleth_mapbox(ls, geojson=geojson, color="neighbourhood", title="Washington D.C. Neighbourhood Map",
                           locations="neighbourhood", featureidkey="properties.neighbourhood",opacity=0.5, color_discrete_sequence=px.colors.qualitative.Light24,
                           center={"lat": 38.9072, "lon": -77.0369},
                           mapbox_style="light", zoom=11)

fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})
fig.show()

ls_d['price'].corr(ls_d['bedrooms'])

0.2036136776639811

Explore numerical attributes¶

# melt listings data for graphing purposes 
melted_ls = pd.melt(ls,id_vars=['id', 'name', 'host_id', 'host_name', 'neighbourhood_group','neighbourhood', 'latitude', 'longitude'], value_vars=ls.select_dtypes(include=np.number).iloc[:, 5:].columns.tolist())

# create dropdown for attributes in melted dataframe
dropdown_attribute = widgets.Dropdown(options = sorted(melted_ls.variable.unique()))

# output
output = widgets.Output()


def view_attribute(attribute):
    
    # clear output for new attribute to be plotted 
    output.clear_output()
    
    # filter df to selected attribute
    filtered = melted_ls[melted_ls['variable'] == attribute].copy()
    
    # filter out outliers 
    filtered = filtered[filtered.value.between(filtered.value.quantile(.10), filtered.value.quantile(.80))]
 
    with output:
        # set token
        px.set_mapbox_access_token("pk.eyJ1IjoibGF3cmVuY2VkIiwiYSI6ImNrODFzZnFnNzA0YmczZW9nNWN4aTFvdngifQ.VlB5-L7owXKEXo8JEePk7w")
        fig = px.scatter_mapbox(filtered.rename({'value':'Listing Price'}, axis=1), lat="latitude", lon="longitude", color="Listing Price", template="gridon", 
                                color_continuous_scale=plotly.colors.diverging.RdYlGn,
                                   opacity=0.4, center={"lat": 38.9072, "lon": -77.0369},
                                   mapbox_style="light", zoom=11)

        fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})
        fig.show()

# change attribute
def dropdown_attribute_handler(change):
    view_attribute(change.new)
    
dropdown_attribute.observe(dropdown_attribute_handler, names="value")
display(dropdown_attribute)
display(output)

How do neighborhoods differ in price?¶

Are certain neighbourhoods generally more expensive than others?¶

Does proximity to attractions play a factor in price?¶

Answer: The median overall for price is around $115, and we can see that neighborhoods that are less popular are at below the median price of the dataset and are colored in red as well.¶

# calculate outliers within neighbourhoods 
stdev = 3.0

zscores = ls[['neighbourhood', 'price']].groupby('neighbourhood').transform(
    lambda group: (group - group.mean()).div(group.std())).abs()

outliers = zscores > stdev

# take out outliers 
ls_nonoutliers = ls[~outliers.any(axis=1)].copy()


# agg all the observations by nieghbourhood to get median and mean 
summary = ls.groupby("neighbourhood").agg({'price':['mean', 'median', 'count']}).reset_index()
summary.columns = ['Neighbourhood', 'Mean Price', 'Median Price', '# of listings']

# strip outliers and redo agg
summary_nonoutliers = ls_nonoutliers.groupby("neighbourhood").agg({'price':['mean', 'count']}).reset_index()
summary_nonoutliers.columns = ['Neighbourhood', 'Mean Price', '# of listings']

# merge 
summary_final = pd.merge(summary, summary_nonoutliers, on='Neighbourhood', suffixes=[" with outliers", " without outliers"])
summary_final["# of outliers"] = summary_final["# of listings with outliers"] - summary_final["# of listings without outliers"] 
summary_final.drop(['# of listings without outliers'], axis=1, inplace=True)

# mean price by nieghborhood 
(
    summary_final
    .sort_values(["Mean Price without outliers"], ascending=False)[['Neighbourhood', 'Mean Price with outliers', 'Mean Price without outliers', 'Median Price', '# of outliers']]
    .style.background_gradient(cmap='RdYlGn', subset=['Mean Price with outliers', 'Median Price', 'Mean Price without outliers']) 
    
          
)

ls_nonoutliers['price'].hist(bins=50)

<matplotlib.axes._subplots.AxesSubplot at 0x1a19a75400>

def format_x(x):
    return "${:.1f}K".format(float(x))


t = (summary_final
    .sort_values(["Mean Price without outliers"], ascending=False)[['Neighbourhood', 'Mean Price without outliers']]
     .rename({'Mean Price without outliers':'Average Price'}, axis=1)
    )

t['Average Price'] = t['Average Price'].apply(lambda x: round(x))

t.style.background_gradient(cmap='RdYlGn', subset=['Average Price'])

# top and bottom twelve nieghborhoods
most_expensive_neighbourhoods = summary_final.sort_values(["Mean Price without outliers"], ascending=False)[:12].Neighbourhood.values.tolist()
cheapest_neighbourhoods= summary_final.sort_values(["Mean Price without outliers"])[:12].Neighbourhood.values.tolist()

top = ls[ls.neighbourhood.isin(most_expensive_neighbourhoods + cheapest_neighbourhoods)]
top['category'] = top['neighbourhood'].apply(lambda x: 'Top 5 Cheapest' if x in cheapest_neighbourhoods else 'Top 5 Most Expensive')


# you save an average of 3x your money by choosing a nieghbourhood from the 5 cheapest rather than the 5 most expsensive 
top.groupby("category").mean()['price'].reset_index().rename({'price':'Average price in Category'}, axis=1)

/Users/ldugom/.local/lib/python3.6/site-packages/ipykernel_launcher.py:6: SettingWithCopyWarning:


A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy

Looking at the 12 most expensive and 12 most inexpensive neighbourhoods visually (39 total)¶

px.set_mapbox_access_token("pk.eyJ1IjoibGF3cmVuY2VkIiwiYSI6ImNrODFzZnFnNzA0YmczZW9nNWN4aTFvdngifQ.VlB5-L7owXKEXo8JEePk7w")
fig = px.scatter_mapbox(top, lat="latitude", lon="longitude", color="category", template="simple_white",center={"lat": 38.895, "lon": -77.024},
                           mapbox_style="basic", zoom=10)

How expensive is the top 5%?¶

Answer: We can see that top listings start at ~ \$600, with a few going all the way to $10,000 ! Georgetown, Capitol, Hill and Downtown/Chinatown seem to pop out the most.¶

# clean price 
ls_d["price"] = ls_d["price"].str[1:].str.replace(",","").astype(float)

# detect outliers (169 total)
outliers_df = ls_d[ls.price.between(ls_d.price.quantile(.95), ls_d.price.quantile(1))].copy().dropna(subset=["beds"])

sns.set_style("dark") 
sns.set_palette("Set3")

fig = px.scatter(outliers_df, y="neighbourhood", x="price", size="bedrooms", color="bedrooms", template="plotly_dark", color_continuous_scale="sunset", width=1500, height=1000)

plotly.io.orca.config.executable = '/Users/ldugom/anaconda3/envs/ds/bin/orca'
fig.write_image("outliers.png", width=1500, height=1000)

ls_d['price'].corr(ls_d['beds'])

0.1922988708297365

From a glance, it looks like the more expensive listings tend to be closer to downtown, but can we assume the farther you move away from the center of DC, the more likely you are to find a cheaper listing?¶

# create bins to eliminate outliers from dominating continuous scale
ls['Price Decile'] = pd.qcut(ls['price'], 10)
ls_d['Price Decile'] = pd.qcut(ls_d['price'], 10)


# set token
px.set_mapbox_access_token("pk.eyJ1IjoibGF3cmVuY2VkIiwiYSI6ImNrODFzZnFnNzA0YmczZW9nNWN4aTFvdngifQ.VlB5-L7owXKEXo8JEePk7w")
fig = px.scatter_mapbox(ls, lat="latitude", lon="longitude", color="Price Decile", template="simple_white", opacity=0.5, 
                         color_discrete_sequence=px.colors.diverging.RdYlGn, category_orders={"Price Decile":sorted(ls['Price Decile'].unique().tolist())}, hover_data=["neighbourhood", "price"],
                           center={"lat": 38.895, "lon": -77.024},
                           mapbox_style="basic", zoom=12)


fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})

fig.write_image("binned_prices.png", height=850, width=1200)

To test our hypothesis about distance to downtown/attractions and being in more of an expensive price range, we'll calculate the distnace to Capital One Arena in Downtown DC as well as the Washington Monument, which is at teh center of a lot of the DC attractions¶

monument = {'lat':38.8872036, 'lon':-77.045968}
arena = {'lat':38.8980942,'lon':-77.0208438}
center = {'lat':38.9072,'lon':-77.0369}
wh = {'lat': 38.8977, 'lon':-77.0365}
from math import radians, cos, sin, asin, sqrt


# Thanks https://stackoverflow.com/questions/4913349/haversine-formula-in-python-bearing-and-distance-between-two-gps-points
    
    
def haversine(lon1, lat1, lon2, lat2):
    """
    Calculate the great circle distance between two points 
    on the earth (specified in decimal degrees)
    """
    # convert decimal degrees to radians 
    lon1, lat1, lon2, lat2 = map(radians, [lon1, lat1, lon2, lat2])

    # haversine formula 
    dlon = lon2 - lon1 
    dlat = lat2 - lat1 
    a = sin(dlat/2)**2 + cos(lat1) * cos(lat2) * sin(dlon/2)**2
    c = 2 * asin(sqrt(a)) 
    r = 3956 
    return c * r

calculate each properties distance to washington monument and capital one arena¶

ls_d["distance_to_monument"] = ls_d[['latitude', 'longitude']] \
    .apply(lambda row: haversine(row['latitude'], row['longitude'], monument.get("lat"), monument.get("lon")), axis=1)


ls_d["distance_to_arena"] = ls_d[['latitude', 'longitude']] \
    .apply(lambda row: haversine(row['latitude'], row['longitude'], arena.get("lat"), arena.get("lon")), axis=1)


ls_d["distance_to_center"] = ls_d[['latitude', 'longitude']] \
    .apply(lambda row: haversine(row['latitude'], row['longitude'], center.get("lat"), center.get("lon")), axis=1)


ls_d["distance_to_whitehouse"] = ls_d[['latitude', 'longitude']] \
    .apply(lambda row: haversine(row['latitude'], row['longitude'], wh.get("lat"), wh.get("lon")), axis=1)

distance_corrs = pd.concat([ls_d.select_dtypes(include=np.number), pd.get_dummies(ls_d['Price Decile'])], axis=1).corr()[["distance_to_monument", "distance_to_arena", "distance_to_center", "distance_to_whitehouse"]].reset_index().dropna()

px.scatter(ls_d[ls_d.review_scores_location.notnull()], x="distance_to_monument", y="review_scores_location", color="review_scores_location", trendline='lowess')

distance_corrs.round(3).sort_values("distance_to_whitehouse").style.background_gradient(low=0.0, high=.10)

Answer: slightly negative assocaition for wh/monumnet distances, location review scores, host listing count. Weak postive correlation betweeen bedrooms, availability and lowest-priced decile of listings.¶

Question 2: What kind of properties are these listings (property type, room type, bedroom count, bathroom count, etc.)¶

Property Types: 44% Apartments, 20% Houses, 15% Townhouses (3 form biggest share)¶

proptypes = (
                ls_d.
                    groupby("property_type").count()["price"]
                    .reset_index().rename({'price':'# of listings'}, axis=1)
)


(
    proptypes
        .sort_values("# of listings", ascending=False)
        .assign(share=proptypes["# of listings"]/ls_d.shape[0])
        .style.background_gradient(cmap='Purples', subset=['share'])
    

)

prop_fig = px.bar(ls_d.groupby(["neighbourhood", "property_type"])['id'].count().reset_index().rename({"id":"listing count"},axis=1).sort_values("listing count",ascending=False),
       title="Property Type by Neighborhood",
       x='neighbourhood', y="listing count", color="property_type", template="plotly_dark", height=900, color_discrete_sequence=plotly.colors.qualitative.Light24)

prop_fig.write_image("property_type.png",height=900, width=1500)

Room Types: We can see that the overwhelming majority of listings are entire homes/apartments ( 71%) , followed by private rooms (~25%).¶

# set token
px.set_mapbox_access_token("pk.eyJ1IjoibGF3cmVuY2VkIiwiYSI6ImNrODFzZnFnNzA0YmczZW9nNWN4aTFvdngifQ.VlB5-L7owXKEXo8JEePk7w")
fig = px.scatter_mapbox(ls_d.dropna(subset=['beds']), lat="latitude", lon="longitude", color="room_type", template="simple_white", size="beds",
                          hover_data=["neighbourhood", "price"],
                           center={"lat": 38.895, "lon": -77.024},
                           mapbox_style="basic", zoom=12)

fig.update_layout(margin={"r":5,"t":0,"l":0,"b":0})

fig.write_image("room_type.png", width=1500, height=900)

roomtypes = (
                ls_d.
                    groupby("room_type").count()["price"]
                    .reset_index().rename({'price':'# of listings'}, axis=1)
)


(
    roomtypes
        .sort_values("# of listings", ascending=False)
        .assign(share=roomtypes["# of listings"]/ls_d.shape[0])
        .style.background_gradient(cmap='Blues', subset=['share'])
    

)

Summary statistics for type of property types: mode of price, bathrooms, bedrooms and beds¶

prop_df = ls_d[['price', 'bathrooms','bedrooms','beds', 'property_type']].copy()

(
    prop_df
        .groupby("property_type").agg(lambda x:x.value_counts().index[0]).reset_index()
        .sort_values("bedrooms",ascending=False)
        .style.background_gradient(cmap='Oranges', subset=["bathrooms","bedrooms", "beds"])
)

Question 3 (Host Analysis): Are there hosts that have multiple listings/have businesses using Aribnb? Do these hosts have good scores and reviews? Do they focus in one or two neighbourhoods? Focus on ceratin price points?¶

Are there hosts that have multiple listings/have businesses using Aribnb?¶

ls_subset = ls_d[['id', 'host_id','host_name', 'host_about', 'host_response_time','host_since', 
      'host_response_rate', 'host_acceptance_rate', 'host_is_superhost', 'host_listings_count', 'host_total_listings_count', 'neighbourhood',
      'latitude', 'longitude', 'room_type', 'property_type', 'bathrooms', 'bedrooms', 'beds', 'price', 'minimum_nights', 'maximum_nights',
      'availability_30', 'availability_60', 'availability_90','availability_365', 'number_of_reviews', 'number_of_reviews_ltm', 
      'review_scores_rating','review_scores_accuracy','review_scores_cleanliness','review_scores_checkin','review_scores_communication','review_scores_location','review_scores_value',
      'reviews_per_month']].copy()

# hosts that have more than one listing 
host_summary = (
    ls_subset
        .groupby(["host_id", "host_name"])
        .count().reset_index()
        .sort_values("id", ascending=False)
        .iloc[:,:3]
        .rename({"id":"# of listings"}, axis=1)
    )

host_summary_multiple = host_summary[host_summary["# of listings"] > 1]

Answer 1: 46% of the listings are owned by hosts with multiple listings. Thes guys are probably running businesses¶

np.sum(host_summary_multiple["# of listings"]) / np.sum(host_summary["# of listings"])

0.4590480274442539

Do these hosts have good scores and reviews?¶

host_reviews = host_summary_multiple.merge(ls_subset, on = ['host_id', 'host_name'])

# weight the scores of the hosts that have more than one listing by "reviews per month"
host_grouped = (
                host_reviews
                    .dropna()
                    .groupby(['host_id', 'host_name'])
                    [['review_scores_rating','review_scores_accuracy','review_scores_cleanliness','review_scores_checkin','review_scores_communication','review_scores_location','review_scores_value',
                          'reviews_per_month']]
)

(
    # collect weighted averages 
    host_grouped
        .apply(lambda x: pd.Series(np.average(x[['review_scores_rating','review_scores_accuracy',
                                                 'review_scores_cleanliness','review_scores_checkin','review_scores_communication',
                                                 'review_scores_location','review_scores_value']], weights=x["reviews_per_month"], axis=0),
                                   ['review_scores_rating','review_scores_accuracy','review_scores_cleanliness','review_scores_checkin',
                                    'review_scores_communication','review_scores_location','review_scores_value']))
    
    # merge in helpful information 
    .merge(host_summary_multiple, on=['host_id', 'host_name'])
    .sort_values("# of listings", ascending=False)
    .style.background_gradient(cmap='RdYlGn',subset=['review_scores_rating','review_scores_accuracy','review_scores_cleanliness','review_scores_checkin',
                                    'review_scores_communication','review_scores_location','review_scores_value'])
    
)

Answer 2: We do not see see too much fluctuation in the scores here, except for a couple bad apples with low accuracy scores as well. It generally looks like the review scores are generally postive for the top hosts and overall¶

ls_d[[col for col in ls_d.columns if "review" in col]].mean()

number_of_reviews              40.574445
number_of_reviews_ltm          14.428127
review_scores_rating           95.106292
review_scores_accuracy          9.727160
review_scores_cleanliness       9.546770
review_scores_checkin           9.837447
review_scores_communication     9.817192
review_scores_location          9.672226
review_scores_value             9.540403
reviews_per_month               1.895331
dtype: float64

rs_d.head()

Do they focus in one or two neighbourhoods? Focus on ceratin price points?¶

# top 100 hosts by # of listings 
top100hosts = host_summary[:100].host_id.values.tolist()

# get mean price of listings acorss neighbourhoods for each host with multiple reviews
prices_neighbourhoods = \
(
    host_reviews[['id', 'host_name', 'host_id', 'price','neighbourhood']]
    .groupby(["host_name","host_id"])
    .agg({'price':'mean', 'neighbourhood':lambda x:x.value_counts().index[0]}, axis=1)
    .reset_index()
    .rename({'price':'mean_price'}, axis=1)
)

multiples = prices_neighbourhoods.merge(host_summary[['host_id','# of listings']], on='host_id').sort_values("# of listings", ascending=False)

host_summary_multiple.head()

Asnwer 3 Part 1: It looks like out of the top 1078 hosts with multiple listings, 825 of them (~77%) are in only one neighourhood¶

h_summary = \
(
    ls[ls.host_id.isin(host_summary_multiple.host_id.unique())]
   .groupby(["host_id", "neighbourhood"])["id"].agg(["count", "sum"]).reset_index()
)

len(h_summary.host_id.value_counts().reset_index().query('host_id > 1')) / len(h_summary.host_id.unique())

0.23469387755102042

multi_neigh = h_summary.host_id.value_counts().reset_index().query('host_id > 1')['index'].unique().tolist()

sns.catplot(height=10, kind='bar',aspect=1.5,
    data = h_summary.groupby("neighbourhood")['count'].sum().reset_index().sort_values("count", ascending=False)
            .rename({'count':'Number of listings', 'neighbourhood':'Neighborhood' }, axis=1),
    y='Neighborhood', x='Number of listings')

<seaborn.axisgrid.FacetGrid at 0x1a2ac34780>

Asnwer 3 Part 2: With the average IQR (of the middle 80%) of hosts with multiple listings being $40, it looks like these "businesses" tend to foucs on certain price points. There is no strong linear relationship between number of listings and IQR eitherm with the correlation around 8%¶

q3 = ls[ls.host_id.isin(multiples.host_id.unique())].groupby(["host_id"])["price"].describe().reset_index()

# filter out outliers
q3_filtered = q3[q3["mean"].between(q3["mean"].quantile(.05), q3["mean"].quantile(.95))]

q3_filtered["IQR"] = q3_filtered["75%"] - q3_filtered["25%"]
                                                                                                                                                      
q3_filtered.sort_values("count", ascending=False).style.background_gradient(cmap="RdYlGn", subset=["25%", "50%", "75%", "IQR"])

/Users/ldugom/.local/lib/python3.6/site-packages/ipykernel_launcher.py:6: SettingWithCopyWarning:


A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy

# what is the mean 
q3_filtered["IQR"].mean()

43.249743326488705

# does IQR tend to go up with # of listings?
q3_filtered["mean"].corr(q3_filtered["IQR"])

0.722507708254381

iqr = px.scatter(q3_filtered.rename({"mean":"Mean Price of listings", "50%":"50th percentile"},axis=1), x="Mean Price of listings", y='IQR', color='50th percentile', trendline='ols', 
                 template="ggplot2", color_continuous_scale=plotly.colors.sequential.Darkmint,
          title="<b>Mean Price vs IQR </b> (hosts with multiple listings)")

iqr.update_yaxes(showgrid=False, tickformat='$')
iqr.update_xaxes(showgrid=False, tickformat='$')

iqr.write_image("iqr.png", width=1200)

Question 4 (Review Analysis): Even though reveiws are subjective, do ceratin neighbourhoods or price ranges reveal any patterns in review sentiment?¶

For this we'll use NLTK's Vader Sentiment Intensity Analyzer. Learn more here: https://medium.com/analytics-vidhya/simplifying-social-media-sentiment-analysis-using-vader-in-python-f9e6ec6fc52f ¶

import nltk
nltk.download('vader_lexicon')

# 'normalized, weighted composite score'
def get_sentiment(sentence):
    
    from nltk.sentiment.vader import SentimentIntensityAnalyzer
    nltk_sentiment = SentimentIntensityAnalyzer()
    score = nltk_sentiment.polarity_scores(sentence)
    return score

[nltk_data] Downloading package vader_lexicon to
[nltk_data]     /Users/ldugom/nltk_data...
[nltk_data]   Package vader_lexicon is already up-to-date!

# make a reviews datframe 
reviews = (
            rs_d
                .rename({'id':'review_id'}, axis=1)
                .merge(ls[['host_id', 'host_name', 'id']], left_on='listing_id', right_on='id')
                .drop("id", axis=1)
)

# attach sentiment scores
reviews_final = pd.concat([reviews, reviews['comments'].astype(str).apply(get_sentiment).apply(pd.Series)], axis=1)

# write for later use

rs.head()

Analyze sentiment scores by host, by neighbourhood, by price range¶

vader_scores = pd.read_csv("/Users/ldugom/sentiment_scores.csv")

vader_merged = vader_scores.merge(ls, left_on=['listing_id'], right_on=['id'] )

vader_summary = vader_merged[['listing_id', 'neighbourhood', 'date', 'neg', 'neu', 'pos', 'compound', 'price']]

Using vader's "normalized" score, it looks all neighboourhoods overall have good reviews, with the bottom 10ish neighbourhoods having relatively lower scores, but all in all, still above average (0-1) scores.¶

vader_groupby = vader_summary.groupby("neighbourhood")["compound"].describe().sort_values(["50%", "count"], ascending=False).reset_index()
vader_groupby.style.background_gradient(cmap="RdYlGn", subset=["25%", "50%", "75%"])

import plotly.express as px

# set token
px.set_mapbox_access_token("pk.eyJ1IjoibGF3cmVuY2VkIiwiYSI6ImNrODFzZnFnNzA0YmczZW9nNWN4aTFvdngifQ.VlB5-L7owXKEXo8JEePk7w")
fig = px.choropleth_mapbox(vader_groupby, geojson=geojson, color="mean", title="Washington D.C. Neighbourhood Map",
                               locations="neighbourhood", featureidkey="properties.neighbourhood",opacity=0.5, color_continuous_scale=px.colors.diverging.Geyser,
                           center={"lat": 38.9072, "lon": -77.0369},
                           mapbox_style="light", zoom=10)

fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})
fig.show()

Does price range uncover patterns in reviews ?¶

vader_merged.groupby("Price Decile")["compound"].describe().reset_index().sort_values("Price Decile").style.background_gradient(cmap="RdYlGn", subset=["mean"])

Answer: Price range does not uncover too much besides the fact that the lowest two priced deciles have relatively lower scores.¶

vader_summary['price'].corr(vader_summary['compound'])

0.03248190402843484

	Neighbourhood	Mean Price with outliers	Mean Price without outliers	Median Price	# of outliers
36	West End, Foggy Bottom, GWU	304.280822	298.625430	150.000000	1
11	Downtown, Chinatown, Penn Quarters, Mount Vernon Square, North Capitol Street	274.610526	247.212766	173.000000	4
17	Georgetown, Burleith/Hillandale	296.938224	235.964844	162.000000	3
31	Southwest Employment Area, Southwest/Waterfront, Fort McNair, Buzzard Point	270.191489	220.862637	189.500000	6
29	Shaw, Logan Circle	290.917910	210.158850	144.500000	9
32	Spring Valley, Palisades, Wesley Heights, Foxhall Crescent, Foxhall Village, Georgetown Reservoir	288.104651	209.250000	122.500000	2
2	Capitol Hill, Lincoln Park	260.604317	207.454545	135.000000	9
20	Howard University, Le Droit Park, Cardozo/Shaw	333.938719	201.375000	130.000000	7
12	Dupont Circle, Connecticut Avenue/K Street	250.692416	195.617232	142.500000	4
5	Cleveland Park, Woodley Park, Massachusetts Avenue Heights, Woodland-Normanstone Terrace	199.610619	179.964286	115.000000	1
35	Union Station, Stanton Park, Kingman Park	220.582380	177.423212	125.000000	21
25	Near Southeast, Navy Yard	177.555556	170.400000	131.000000	1
4	Cathedral Heights, McLean Gardens, Glover Park	204.808824	164.541985	99.500000	5
18	Hawthorne, Barnaby Woods, Chevy Chase	235.377049	164.389831	100.000000	2
22	Kalorama Heights, Adams Morgan, Lanier Heights	174.108108	144.592500	119.000000	7
26	North Cleveland Park, Forest Hills, Van Ness	198.028169	143.432836	110.000000	4
14	Edgewood, Bloomingdale, Truxton Circle, Eckington	168.550484	137.110638	100.000000	18
15	Fairfax Village, Naylor Gardens, Hillcrest, Summit Park	128.636364	128.636364	99.000000	0
7	Columbia Heights, Mt. Pleasant, Pleasant Plains, Park View	148.452915	128.066514	99.000000	20
8	Congress Heights, Bellevue, Washington Highlands	166.463415	126.250000	92.000000	2
37	Woodland/Fort Stanton, Garfield Heights, Knox Hill	120.555556	120.555556	69.000000	0
28	River Terrace, Benning, Greenway, Dupont Park	179.055556	120.169811	70.000000	1
19	Historic Anacostia	143.360656	119.406780	75.000000	2
0	Brightwood Park, Crestwood, Petworth	144.827977	118.123791	88.000000	12
34	Twining, Fairlawn, Randle Highlands, Penn Branch, Fort Davis Park, Fort Dupont	140.739496	117.504274	80.000000	2
16	Friendship Heights, American University Park, Tenleytown	136.488372	112.506024	99.000000	3
21	Ivy City, Arboretum, Trinidad, Carver Langston	127.266917	110.702290	95.000000	4
6	Colonial Village, Shepherd Park, North Portal Estates	127.975000	109.461538	86.000000	1
10	Douglas, Shipley Terrace	143.709677	106.833333	65.000000	1
23	Lamont Riggs, Queens Chapel, Fort Totten, Pleasant Hill	131.128205	102.086207	75.000000	1
3	Capitol View, Marshall Heights, Benning Heights	118.040000	102.067568	70.000000	1
1	Brookland, Brentwood, Langdon	118.192771	96.822086	79.500000	3
30	Sheridan, Barry Farm, Buena Vista	136.478261	95.066667	72.000000	1
38	Woodridge, Fort Lincoln, Gateway	111.846154	93.190476	80.000000	2
27	North Michigan Park, Michigan Park, University Heights	110.239583	89.180851	75.000000	2
33	Takoma, Brightwood, Manor Park	96.852041	85.927461	68.000000	3
13	Eastland Gardens, Kenilworth	108.428571	83.307692	70.000000	1
24	Mayfair, Hillbrook, Mahaning Heights	94.491525	83.189655	65.000000	1
9	Deanwood, Burrville, Grant Park, Lincoln Heights, Fairmont Heights	92.765957	75.333333	75.000000	2

	Neighbourhood	Average Price
36	West End, Foggy Bottom, GWU	299
11	Downtown, Chinatown, Penn Quarters, Mount Vernon Square, North Capitol Street	247
17	Georgetown, Burleith/Hillandale	236
31	Southwest Employment Area, Southwest/Waterfront, Fort McNair, Buzzard Point	221
29	Shaw, Logan Circle	210
32	Spring Valley, Palisades, Wesley Heights, Foxhall Crescent, Foxhall Village, Georgetown Reservoir	209
2	Capitol Hill, Lincoln Park	207
20	Howard University, Le Droit Park, Cardozo/Shaw	201
12	Dupont Circle, Connecticut Avenue/K Street	196
5	Cleveland Park, Woodley Park, Massachusetts Avenue Heights, Woodland-Normanstone Terrace	180
35	Union Station, Stanton Park, Kingman Park	177
25	Near Southeast, Navy Yard	170
4	Cathedral Heights, McLean Gardens, Glover Park	165
18	Hawthorne, Barnaby Woods, Chevy Chase	164
22	Kalorama Heights, Adams Morgan, Lanier Heights	145
26	North Cleveland Park, Forest Hills, Van Ness	143
14	Edgewood, Bloomingdale, Truxton Circle, Eckington	137
15	Fairfax Village, Naylor Gardens, Hillcrest, Summit Park	129
7	Columbia Heights, Mt. Pleasant, Pleasant Plains, Park View	128
8	Congress Heights, Bellevue, Washington Highlands	126
37	Woodland/Fort Stanton, Garfield Heights, Knox Hill	121
28	River Terrace, Benning, Greenway, Dupont Park	120
19	Historic Anacostia	119
0	Brightwood Park, Crestwood, Petworth	118
34	Twining, Fairlawn, Randle Highlands, Penn Branch, Fort Davis Park, Fort Dupont	118
16	Friendship Heights, American University Park, Tenleytown	113
21	Ivy City, Arboretum, Trinidad, Carver Langston	111
6	Colonial Village, Shepherd Park, North Portal Estates	109
10	Douglas, Shipley Terrace	107
23	Lamont Riggs, Queens Chapel, Fort Totten, Pleasant Hill	102
3	Capitol View, Marshall Heights, Benning Heights	102
1	Brookland, Brentwood, Langdon	97
30	Sheridan, Barry Farm, Buena Vista	95
38	Woodridge, Fort Lincoln, Gateway	93
27	North Michigan Park, Michigan Park, University Heights	89
33	Takoma, Brightwood, Manor Park	86
13	Eastland Gardens, Kenilworth	83
24	Mayfair, Hillbrook, Mahaning Heights	83
9	Deanwood, Burrville, Grant Park, Lincoln Heights, Fairmont Heights	75

	index	distance_to_monument	distance_to_arena	distance_to_center	distance_to_whitehouse
9	latitude	-0.375000	-0.158000	-0.373000	-0.354000
37	review_scores_location	-0.280000	-0.137000	-0.256000	-0.255000
40	calculated_host_listings_count_entire_homes	-0.163000	-0.088000	-0.150000	-0.153000
39	calculated_host_listings_count	-0.156000	-0.087000	-0.144000	-0.147000
6	host_listings_count	-0.159000	-0.074000	-0.139000	-0.143000
7	host_total_listings_count	-0.159000	-0.074000	-0.139000	-0.143000
15	square_feet	-0.133000	-0.103000	-0.139000	-0.141000
24	minimum_nights_avg_ntm	-0.115000	-0.041000	-0.096000	-0.098000
21	maximum_minimum_nights	-0.115000	-0.042000	-0.096000	-0.098000
54	(135.0, 165.0]	-0.074000	-0.057000	-0.073000	-0.074000
16	price	-0.060000	-0.038000	-0.054000	-0.056000
57	(350.0, 10000.0]	-0.064000	-0.040000	-0.052000	-0.055000
20	minimum_minimum_nights	-0.062000	-0.017000	-0.053000	-0.054000
55	(165.0, 215.0]	-0.052000	-0.041000	-0.051000	-0.053000
56	(215.0, 350.0]	-0.036000	-0.043000	-0.036000	-0.038000
53	(115.0, 135.0]	-0.037000	-0.020000	-0.033000	-0.033000
52	(99.0, 115.0]	-0.034000	-0.023000	-0.032000	-0.032000
18	minimum_nights	-0.042000	0.002000	-0.032000	-0.031000
51	(85.0, 99.0]	-0.015000	0.016000	-0.012000	-0.009000
33	review_scores_accuracy	0.004000	-0.011000	-0.002000	-0.002000
1	scrape_id	0.000000	-0.000000	0.000000	-0.000000
19	maximum_nights	0.005000	-0.003000	0.004000	0.003000
22	minimum_maximum_nights	0.005000	-0.003000	0.004000	0.003000
23	maximum_maximum_nights	0.005000	-0.003000	0.004000	0.003000
25	maximum_nights_avg_ntm	0.005000	-0.003000	0.004000	0.003000
32	review_scores_rating	0.007000	-0.002000	0.005000	0.005000
30	number_of_reviews	0.026000	-0.015000	0.021000	0.020000
29	availability_365	0.021000	0.005000	0.023000	0.022000
41	calculated_host_listings_count_private_rooms	0.028000	0.017000	0.026000	0.026000
12	bathrooms	0.037000	-0.003000	0.030000	0.029000
34	review_scores_cleanliness	0.034000	0.016000	0.031000	0.031000
42	calculated_host_listings_count_shared_rooms	0.052000	-0.025000	0.036000	0.034000
50	(70.0, 85.0]	0.045000	0.023000	0.037000	0.039000
36	review_scores_communication	0.045000	0.026000	0.041000	0.042000
38	review_scores_value	0.048000	0.019000	0.041000	0.042000
31	number_of_reviews_ltm	0.054000	0.001000	0.047000	0.046000
17	guests_included	0.063000	-0.017000	0.049000	0.047000
35	review_scores_checkin	0.050000	0.044000	0.048000	0.050000
0	id	0.065000	0.019000	0.059000	0.056000
43	reviews_per_month	0.069000	0.002000	0.060000	0.058000
11	accommodates	0.081000	-0.011000	0.067000	0.064000
14	beds	0.087000	0.019000	0.079000	0.077000
5	host_id	0.093000	0.032000	0.083000	0.082000
49	(53.0, 70.0]	0.088000	0.066000	0.083000	0.085000
27	availability_60	0.088000	0.047000	0.086000	0.085000
28	availability_90	0.091000	0.043000	0.087000	0.086000
13	bedrooms	0.100000	0.022000	0.088000	0.087000
26	availability_30	0.102000	0.052000	0.098000	0.098000
48	(-0.001, 53.0]	0.175000	0.118000	0.165000	0.167000
10	longitude	0.847000	0.231000	0.717000	0.708000
45	distance_to_arena	0.625000	1.000000	0.779000	0.793000
44	distance_to_monument	1.000000	0.625000	0.967000	0.964000
46	distance_to_center	0.967000	0.779000	1.000000	0.999000
47	distance_to_whitehouse	0.964000	0.793000	0.999000	1.000000

	property_type	# of listings	share
1	Apartment	4127	0.442384
15	House	1933	0.207203
21	Townhouse	1402	0.150284
8	Condominium	764	0.081895
11	Guest suite	547	0.058634
19	Serviced apartment	241	0.025833
16	Loft	69	0.007396
12	Guesthouse	53	0.005681
3	Bed and breakfast	47	0.005038
5	Boutique hotel	41	0.004395
13	Hostel	34	0.003645
17	Other	20	0.002144
6	Bungalow	17	0.001822
14	Hotel	9	0.000965
22	Villa	8	0.000858
4	Boat	4	0.000429
20	Tiny house	4	0.000429
0	Aparthotel	3	0.000322
2	Barn	2	0.000214
10	Dome house	1	0.000107
9	Cottage	1	0.000107
7	Camper/RV	1	0.000107
18	Resort	1	0.000107

	room_type	# of listings	share
0	Entire home/apt	6652	0.713045
2	Private room	2414	0.258763
3	Shared room	219	0.023475
1	Hotel room	44	0.004716

	property_type	price	bathrooms	bedrooms	beds
10	Dome house	300.000000	4.500000	4.000000	4.000000
2	Barn	275.000000	3.000000	3.000000	3.000000
4	Boat	99.000000	1.000000	2.000000	5.000000
0	Aparthotel	253.000000	1.000000	1.000000	1.000000
13	Hostel	7000.000000	1.000000	1.000000	6.000000
21	Townhouse	150.000000	1.000000	1.000000	1.000000
19	Serviced apartment	369.000000	1.000000	1.000000	1.000000
18	Resort	500.000000	1.000000	1.000000	2.000000
17	Other	272.000000	1.000000	1.000000	1.000000
16	Loft	400.000000	1.000000	1.000000	2.000000
15	House	75.000000	1.000000	1.000000	1.000000
14	Hotel	40.000000	1.000000	1.000000	2.000000
11	Guest suite	100.000000	1.000000	1.000000	1.000000
12	Guesthouse	120.000000	1.000000	1.000000	1.000000
1	Apartment	100.000000	1.000000	1.000000	1.000000
9	Cottage	100.000000	1.000000	1.000000	1.000000
8	Condominium	150.000000	1.000000	1.000000	1.000000
6	Bungalow	100.000000	1.000000	1.000000	1.000000
5	Boutique hotel	100.000000	1.000000	1.000000	1.000000
3	Bed and breakfast	189.000000	1.000000	1.000000	1.000000
22	Villa	80.000000	1.000000	1.000000	1.000000
7	Camper/RV	50.000000	1.000000	0.000000	5.000000
20	Tiny house	175.000000	1.000000	0.000000	1.000000

	listing_id	id	date	reviewer_id	reviewer_name	comments
0	3362	662	2009-01-21	6917	Robyn	Ayeh is a gracious and accommodating host. He...
1	3362	1099	2009-03-26	10492	Brian	I stayed at Ayeh's place in March, and it was ...
2	3362	6300	2009-07-28	22668	Anthony	My 14 y.o. niece, her best friend and I stayed...
3	3362	8387	2009-08-27	30379	Lina	We enjoyed the time with Ayeh in Washington ve...
4	3362	8451	2009-08-28	33059	Nicole	We also enjoyed our stay in DC while at Ayeh's...

	host_id	host_name	# of listings
3649	48005494	Zeus	186
4853	107434423	Blueground	165
3606	46630199	Home Sweet City	64
3346	39930655	Nicole &	62
31	17633	Charlotte (& Anna-Regina)	54

	neighbourhood	count	mean	std	min	25%	50%	75%	max
0	Colonial Village, Shepherd Park, North Portal Estates	1030.000000	0.830860	0.256456	-0.831600	0.826800	0.928450	0.966950	0.997800
1	Capitol Hill, Lincoln Park	42683.000000	0.819560	0.278853	-0.991200	0.812600	0.923100	0.965500	0.999500
2	Friendship Heights, American University Park, Tenleytown	5232.000000	0.824316	0.260883	-0.988800	0.807400	0.920850	0.963900	0.999000
3	North Cleveland Park, Forest Hills, Van Ness	898.000000	0.812953	0.284073	-0.978100	0.804550	0.920250	0.964275	0.996500
4	Hawthorne, Barnaby Woods, Chevy Chase	2204.000000	0.811768	0.273531	-0.975000	0.796300	0.920050	0.965100	0.998200
5	Cathedral Heights, McLean Gardens, Glover Park	3653.000000	0.812637	0.278961	-0.977600	0.806600	0.919500	0.963200	0.997900
6	Howard University, Le Droit Park, Cardozo/Shaw	13732.000000	0.808807	0.285925	-0.991900	0.799575	0.916400	0.961200	0.999200
7	Near Southeast, Navy Yard	1125.000000	0.816439	0.267780	-0.991000	0.795100	0.916100	0.967000	0.998700
8	Columbia Heights, Mt. Pleasant, Pleasant Plains, Park View	36651.000000	0.802207	0.291870	-0.990500	0.789700	0.915000	0.961100	0.999000
9	Union Station, Stanton Park, Kingman Park	45557.000000	0.800956	0.295340	-0.992500	0.790600	0.914400	0.961800	0.999300
10	Cleveland Park, Woodley Park, Massachusetts Avenue Heights, Woodland-Normanstone Terrace	4119.000000	0.812092	0.265351	-0.991900	0.795500	0.913100	0.960100	0.999100
11	Dupont Circle, Connecticut Avenue/K Street	27062.000000	0.812291	0.267600	-0.994400	0.795100	0.911500	0.958700	0.998700
12	North Michigan Park, Michigan Park, University Heights	3354.000000	0.785971	0.318991	-0.980300	0.777800	0.911450	0.961100	0.998500
13	Kalorama Heights, Adams Morgan, Lanier Heights	13411.000000	0.811656	0.271185	-0.990400	0.795950	0.911400	0.959500	0.999000
14	Brightwood Park, Crestwood, Petworth	18675.000000	0.796668	0.294904	-0.994200	0.777800	0.909900	0.960600	0.999400
15	Shaw, Logan Circle	32336.000000	0.800602	0.287426	-0.996500	0.784500	0.909700	0.959300	0.999600
16	Takoma, Brightwood, Manor Park	5271.000000	0.788944	0.300338	-0.992400	0.770700	0.907500	0.958700	0.998100
17	Spring Valley, Palisades, Wesley Heights, Foxhall Crescent, Foxhall Village, Georgetown Reservoir	2472.000000	0.783361	0.305192	-0.938100	0.762000	0.906200	0.959525	0.998100
18	Edgewood, Bloomingdale, Truxton Circle, Eckington	37290.000000	0.779893	0.319276	-0.999000	0.763400	0.906000	0.959000	0.999200
19	Georgetown, Burleith/Hillandale	12244.000000	0.795383	0.288874	-0.996300	0.777700	0.904200	0.957525	0.998800
20	Eastland Gardens, Kenilworth	748.000000	0.756188	0.361569	-0.987300	0.750600	0.899900	0.956000	0.999100
21	Brookland, Brentwood, Langdon	7616.000000	0.769173	0.322171	-0.993200	0.733900	0.899600	0.957100	0.997900
22	West End, Foggy Bottom, GWU	9908.000000	0.798010	0.269734	-0.992800	0.778300	0.897900	0.951600	0.998500
23	Douglas, Shipley Terrace	1510.000000	0.770736	0.328035	-0.995300	0.741200	0.896800	0.958075	0.997200
24	Capitol View, Marshall Heights, Benning Heights	2833.000000	0.752043	0.347429	-0.984300	0.705600	0.895700	0.955800	0.998300
25	Lamont Riggs, Queens Chapel, Fort Totten, Pleasant Hill	3306.000000	0.769149	0.318053	-0.991900	0.739100	0.893100	0.955900	0.997900
26	Fairfax Village, Naylor Gardens, Hillcrest, Summit Park	596.000000	0.780133	0.294929	-0.976300	0.753675	0.892650	0.948000	0.997900
27	Downtown, Chinatown, Penn Quarters, Mount Vernon Square, North Capitol Street	9667.000000	0.765704	0.316928	-0.986400	0.723850	0.890800	0.951000	0.998200
28	Ivy City, Arboretum, Trinidad, Carver Langston	10915.000000	0.739560	0.352861	-0.996100	0.687200	0.883300	0.951500	0.998500
29	Southwest Employment Area, Southwest/Waterfront, Fort McNair, Buzzard Point	5484.000000	0.744949	0.334101	-0.992600	0.669700	0.880500	0.947750	0.999300
30	Mayfair, Hillbrook, Mahaning Heights	1331.000000	0.734551	0.353913	-0.989800	0.678950	0.880300	0.953100	0.995000
31	Woodridge, Fort Lincoln, Gateway	2277.000000	0.750099	0.322390	-0.991400	0.680800	0.877700	0.948900	0.997500
32	Historic Anacostia	2381.000000	0.721508	0.362680	-0.982300	0.636900	0.868700	0.949800	0.998500
33	Congress Heights, Bellevue, Washington Highlands	2640.000000	0.730369	0.329258	-0.988200	0.636900	0.862050	0.940825	0.998900
34	Twining, Fairlawn, Randle Highlands, Penn Branch, Fort Davis Park, Fort Dupont	3327.000000	0.704898	0.366367	-0.989900	0.624900	0.861900	0.944850	0.998400
35	Deanwood, Burrville, Grant Park, Lincoln Heights, Fairmont Heights	865.000000	0.732391	0.322536	-0.984400	0.658800	0.861000	0.936000	0.998500
36	Sheridan, Barry Farm, Buena Vista	1385.000000	0.688582	0.397542	-0.976900	0.624900	0.858800	0.936000	0.997100
37	River Terrace, Benning, Greenway, Dupont Park	2488.000000	0.694554	0.377851	-0.989200	0.624900	0.851900	0.944000	0.997000
38	Woodland/Fort Stanton, Garfield Heights, Knox Hill	243.000000	0.719297	0.328980	-0.916800	0.647850	0.843900	0.918450	0.992500

	Price Decile	count	mean	std	min	25%	50%	75%	max
0	(-0.001, 53.0]	32654.000000	0.727789	0.348752	-0.999000	0.658800	0.871600	0.946700	0.999200
1	(53.0, 70.0]	45919.000000	0.767025	0.326381	-0.995300	0.733900	0.897900	0.956200	0.999600
2	(70.0, 85.0]	48096.000000	0.791076	0.305517	-0.993300	0.778300	0.910100	0.960000	0.999300
3	(85.0, 99.0]	49222.000000	0.803917	0.292464	-0.996100	0.791875	0.915100	0.961800	0.999400
4	(99.0, 115.0]	40111.000000	0.802190	0.287598	-0.996500	0.788700	0.910800	0.960000	0.999400
5	(115.0, 135.0]	49168.000000	0.811322	0.276231	-0.996300	0.796000	0.913500	0.961000	0.999200
6	(135.0, 165.0]	44409.000000	0.818933	0.269735	-0.991200	0.806700	0.918600	0.962600	0.999100
7	(165.0, 215.0]	32232.000000	0.798647	0.290408	-0.992800	0.784000	0.909700	0.959500	0.999000
8	(215.0, 350.0]	26185.000000	0.806802	0.286006	-0.991700	0.790900	0.915900	0.962700	0.999500
9	(350.0, 10000.0]	10523.000000	0.811074	0.276294	-0.987400	0.800750	0.915100	0.961550	0.998200

	category	Average price in Category
0	Top 5 Cheapest	113.399160
1	Top 5 Most Expensive	265.273309

	host_id	host_name	review_scores_rating	review_scores_accuracy	review_scores_cleanliness	review_scores_checkin	review_scores_communication	review_scores_location	review_scores_value	# of listings
425	48005494	Zeus	97.086214	9.805689	9.760613	9.821444	9.677899	9.865208	9.032823	186
521	107434423	Blueground	95.125448	9.784946	9.412186	8.716846	9.419355	9.261649	8.172043	165
420	46630199	Home Sweet City	96.511743	9.911369	9.966854	9.951760	9.960031	9.873694	9.523818	64
398	39930655	Nicole &	95.904617	9.780509	9.777721	9.820006	9.926161	9.986314	9.642124	62
11	17633	Charlotte (& Anna-Regina)	94.839660	9.494901	9.703966	9.705382	9.411048	9.900000	9.424363	54
195	8008783	Cliff	92.767093	9.592485	9.399380	9.766705	9.016076	9.922913	9.180322	51
293	19131071	Agnes	60.000000	2.000000	4.000000	10.000000	10.000000	10.000000	8.000000	29
36	487806	Stay BnB DC	92.772523	9.366132	9.343984	9.910601	9.826047	9.934710	9.112239	27
470	69203193	Leyla And Matt	92.241787	9.305314	9.308213	9.752899	9.689372	9.863768	9.111111	27
30	315148	John	92.733882	9.153635	8.315501	9.687243	9.543210	9.382716	9.117970	25
274	16644446	Corporate	94.028037	9.651090	9.383178	9.666667	9.780374	9.954829	9.401869	24
219	10292686	Brian	97.361546	9.967801	9.935603	10.000000	10.000000	10.000000	9.728611	23
145	4962900	Stay Alfred	91.311659	9.587444	9.313901	8.961883	9.199552	9.802691	9.260090	23
499	95459395	Bluebird	89.000000	9.586777	9.586777	9.380165	8.818182	9.793388	8.446281	22
529	122382567	Evolve	91.048193	9.693976	9.024096	10.000000	10.000000	9.903614	9.296386	20
157	5487930	Mark	93.680451	9.501880	8.832707	9.659774	9.849624	10.000000	9.214286	19
384	35549750	Morgan	95.729515	9.823389	9.125961	9.982233	10.000000	9.919915	9.848846	19
247	13817352	Pam	93.914622	9.565260	9.847890	9.942427	9.976120	9.981354	9.038927	17
594	219493221	Sonder (D.C.)	95.832407	9.851852	9.783333	9.929167	9.944444	9.214815	9.459259	17
567	177578477	Jarrick, Ferris, Rita	93.862314	9.700497	9.367637	10.000000	9.771469	9.239177	9.403123	16
120	3672670	Dennis	88.844437	9.080344	8.537940	9.988924	9.745412	8.934865	8.861299	16
353	26180779	Will	91.576214	9.456951	8.960134	9.926633	9.960134	9.879732	9.089112	16
76	1864688	Blü Properties	96.710602	9.932665	9.561605	10.000000	10.000000	10.000000	9.524037	15
316	21493494	RJ Management LLC	96.044187	9.959075	9.645611	10.000000	10.000000	9.914887	9.654211	15
613	277396425	Convention Center Exclusive Flats	87.305365	8.840440	9.685007	7.903714	9.832187	9.059147	9.242091	15
85	2131753	Shannon	93.876289	9.807560	9.285223	9.470790	9.704467	10.000000	9.278351	12
209	9419684	Mike	90.865052	8.141869	9.837370	8.269896	8.346021	10.000000	9.730104	12
27	138459	L	99.720000	10.000000	10.000000	10.000000	9.773333	10.000000	9.946667	12
337	23714240	Nana	92.221102	9.617953	9.612913	9.937008	9.748031	9.547717	9.120945	11
419	46582948	Huxley	97.398793	9.934449	9.889194	9.895228	10.000000	9.909490	9.624520	11
554	158057505	Namastay	92.127530	9.440536	9.382338	9.382338	9.464828	9.614119	9.000000	11
595	219971307	Abrahim	94.414663	9.788462	9.348558	9.836538	9.902644	9.539663	9.555288	11
254	14692970	P	91.480916	9.366412	9.811069	9.009542	9.303435	9.835878	9.000000	10
113	3448315	Allen	94.588939	9.704036	9.562033	9.977578	9.970105	9.696562	9.623318	10
607	250464525	Morgan	88.883529	9.284706	8.822353	9.075294	9.652941	9.327059	9.652941	10
438	54126082	Jason	93.904282	9.908480	9.673384	9.733837	9.487825	9.753988	9.422334	10
251	14275599	Heather	98.277384	10.000000	10.000000	10.000000	10.000000	10.000000	10.000000	9
503	98799358	Jacob	97.469449	10.000000	10.000000	10.000000	10.000000	9.233979	9.628912	9
615	288033145	Linas	92.375000	10.000000	9.750000	9.500000	9.125000	9.750000	9.750000	9
586	205105773	Anita	96.733333	9.764198	9.730864	10.000000	10.000000	10.000000	9.764198	9

	host_id	count	mean	std	min	25%	50%	75%	max	IQR
639	48005494	186.000000	160.505376	46.191639	60.000000	128.250000	149.000000	179.000000	380.000000	50.750000
831	107434423	165.000000	148.103030	32.826340	64.000000	123.000000	147.000000	174.000000	231.000000	51.000000
633	46630199	64.000000	175.015625	73.073456	90.000000	129.000000	149.000000	199.000000	450.000000	70.000000
579	39930655	62.000000	192.419355	81.542415	71.000000	135.000000	165.000000	248.750000	425.000000	113.750000
13	17633	54.000000	200.425926	101.502784	99.000000	139.000000	177.000000	230.000000	739.000000	91.000000
261	8008783	51.000000	367.588235	219.082466	120.000000	195.000000	299.000000	497.500000	1200.000000	302.500000
521	30283594	40.000000	183.175000	36.960747	129.000000	156.500000	179.000000	201.500000	289.000000	45.000000
408	19131071	29.000000	49.758621	14.846008	28.000000	40.000000	50.000000	55.000000	99.000000	15.000000
716	69203193	27.000000	107.629630	23.125324	90.000000	99.000000	100.000000	100.000000	200.000000	1.000000
43	487806	27.000000	78.111111	17.720769	58.000000	65.000000	75.000000	85.000000	135.000000	20.000000
380	16644446	24.000000	306.083333	68.364698	199.000000	271.500000	299.000000	369.000000	399.000000	97.500000
497	26389480	24.000000	61.083333	25.832094	35.000000	37.000000	62.000000	79.000000	105.000000	42.000000
300	10292686	23.000000	183.695652	55.702485	125.000000	160.000000	165.000000	190.000000	375.000000	30.000000
1028	257036056	23.000000	96.130435	68.095724	40.000000	54.000000	70.000000	101.000000	300.000000	47.000000
1029	257055626	23.000000	50.217391	26.392073	35.000000	40.000000	45.000000	45.000000	160.000000	5.000000
787	95459395	22.000000	334.545455	172.207441	170.000000	199.000000	252.000000	426.750000	800.000000	227.750000
853	122382567	20.000000	349.250000	228.215244	109.000000	231.000000	282.500000	377.000000	1134.000000	146.000000
847	117181690	20.000000	212.800000	39.651906	79.000000	200.000000	220.000000	222.000000	299.000000	22.000000
206	5487930	19.000000	96.526316	29.747770	60.000000	81.500000	95.000000	102.500000	200.000000	21.000000
871	133109918	17.000000	165.000000	38.860327	106.000000	130.000000	171.000000	192.000000	227.000000	62.000000
343	13817352	17.000000	258.235294	129.939183	108.000000	175.000000	250.000000	300.000000	569.000000	125.000000
999	219493221	17.000000	229.882353	75.904448	125.000000	174.000000	220.000000	255.000000	415.000000	81.000000
153	3672670	16.000000	95.625000	71.066049	48.000000	60.000000	70.000000	74.500000	300.000000	14.500000
593	41656215	16.000000	240.937500	211.532021	59.000000	94.250000	144.500000	297.000000	795.000000	202.750000
939	177578477	16.000000	45.687500	22.458016	30.000000	30.000000	34.000000	52.500000	95.000000	22.500000
1045	276944253	15.000000	116.933333	52.303328	59.000000	74.500000	90.000000	169.000000	200.000000	94.500000
440	21493494	15.000000	188.600000	104.777860	121.000000	125.000000	139.000000	219.000000	429.000000	94.000000
98	1864688	15.000000	86.066667	39.290978	50.000000	50.000000	75.000000	115.500000	155.000000	65.500000
1040	268622314	13.000000	100.000000	0.000000	100.000000	100.000000	100.000000	100.000000	100.000000	0.000000
32	138459	12.000000	167.416667	35.369371	90.000000	144.000000	184.000000	189.000000	209.000000	45.000000
286	9419684	12.000000	290.916667	57.809182	215.000000	272.000000	272.000000	308.500000	429.000000	36.500000
107	2131753	12.000000	133.000000	26.724180	115.000000	119.000000	122.000000	130.500000	209.000000	11.500000
1000	219971307	11.000000	149.909091	122.512411	59.000000	68.500000	80.000000	199.000000	399.000000	130.500000
909	158057505	11.000000	125.272727	28.474869	95.000000	103.500000	116.000000	155.000000	172.000000	51.500000
471	23714240	11.000000	157.909091	57.310478	82.000000	106.500000	181.000000	194.500000	250.000000	88.000000
1055	294545484	11.000000	59.818182	7.250078	43.000000	57.000000	60.000000	64.000000	69.000000	7.000000
632	46582948	11.000000	238.272727	59.691023	180.000000	195.000000	210.000000	274.000000	350.000000	79.000000
666	54126082	10.000000	206.300000	127.654960	75.000000	138.750000	176.500000	200.000000	500.000000	61.250000
353	14692970	10.000000	125.700000	43.787492	91.000000	99.500000	103.000000	130.000000	210.000000	30.500000
145	3448315	10.000000	271.800000	170.596209	97.000000	119.000000	200.000000	436.250000	495.000000	317.250000